Alfa calibration scans, rotation angle
Alfa was reinstalled in the dome on 30oct09 after
being worked on in the lab. Spider scans (heiles
calibration scans) were taken with alfa on 02nov09 using beam 0 and
03nov09 using beams 0,2,3,4,5 to test its
- The alfa rotation angle was set to 0 degrees.
- Spare bias boards from the lab were used to drive
beam0, beams 2,3,4,5 were driven by the standard cards.
- The 100 MHz filter (1390 to 1490) was used.
- The wapps took data with 100 MHz bw centered at 1440 MHz.
- On 02nov09 beam 0 looked at 4 sources. On 03nov09 beams 0,2,3,4,5
alternated on various sources.
Calibration run results:
The plots show the
calibration results (.ps) (.pdf):
- Page 1: the az/za coverage. Colors specify the beam that was used.
- Page 2: gain,Tsys,SEFD, avgBeam width. This used the standard cal
values. Colors show the different beams
- beam 2 has large tsys and large sefd. Since the sefd is large,
this is a problem with Tsys and not the cals.
- Page 3: Gain,Tsys,SEFD for beam 2 with polA,B separated.
- polB is ok, polA is bad
- for polA the gain matches polB. The tsys and sefd are different
so the problem is with Tsys polb and it is not a problem with the cals.
- Page 4: coma, 1st side lobe height, beam efficiency (main beam)
and (main beam + first side lobe).
- Page 5: Pointing error. beam
2 had large za errors. beam
3 had large az errors.
Fitting for rotation angle error:
Part of the pointing error could come from the alfa
rotation angle being off. To test for this :
Plots showing the
pointing error vs rotation angle offset (.ps) (.pdf):
- Compute the change in the pointing error if the alfa rotation
error was off by -5 to +5 degrees in .1 degree steps. Do this for each
- For each of the .1 degree steps recompute what the pointing error
would have been for each measured position.
- Plot the rms error and mean error for these newly computed
series. Do this separately for beam 0 (which is not affected by
rotation) and the other beams.
- Top: rms pointing error vs rotation angle offset.
- For each offset (-5 to .5 in .1 degree steps) the rms pointing
error was recomputed
- + beam 0, * beams 2,3,4,5
- Black is the total
rms (rmsAz, rmsZa added in quadrature)
- Red: Az rms pointing
- Green: Za rms pointing
- The total rms minimizes at a rotation offset of 2.0 degrees. It
is about 4 asecs less than the value with not rotation offset (what is
currently being used).
- beam 0 totRms is 8 asecs, nonBm0 0 rotation=18.7 asecs, nonBm0
- The az and za minimums occur and different rotation offsets.
This may be from the limited number of measurements we took and our
excluding beams 1,6.
- Bottom: mean pointing error vs rotation angle offset
- Black tot Mean error. az and za added in quadrature.
- green: mean za error,
- red : mean az error.
- minimum mean error occurred at 1.4 degrees. This is close to
the 1.55 degrees measured back in 2004.
- At 1.4 degree offset, the beam 0 and non beam 0 offsets are the
- The gain,sefd, cals look ok, except for bm 2 polA which has a
Tsys around 120K.
- pointing errors:
- rotating by + 2 degrees reduces the non beam 0 rms from 18.7 to
|rms Error asecs
|mean error asecs
- The az,za coverage was not uniform for all the beams (there was
no coverage for beams 1,6).